Tuberculosis causes almost 3 million deaths per year worldwide. A vaccine effective against tuberculosis is essential to control or elimination of this disease. Great strides have been made in our understanding of the immune responses important in protection against this disease, and this knowledge is fundamental in the design of a vaccine. A non-human primate model for tuberculosis would provide answers to many questions about immune responses to M. tuberculosis as well as provide an animal model for testing vaccines and drugs. Results obtained in such a model may be relevant to human studies. Here we propose three specific aims: 1) establishment of a non-human primate model for tuberculosis; 2) assessment of immunologic correlates of tuberculosis in a non-human primate model; and 3) the effect of coinfection with M. tuberculosis and SIV on progression to AIDS, as well as reactivation of latent tuberculosis. Monkeys with tuberculosis can be quite infectious to humans and to monkeys. A primate Biosafety Level 3 facility will be outfitted so that appropriate and safe techniques are used in these studies. We will compare rhesus macaque and cynomolgus monkeys with respect to the ability to control a low dose (10 CFU, delivered to the lungs via bronchoscope) virulent M. tuberculosis infection, to determine whether a subset of animals of either species can control the infection in a chronic (or latent) state. Establishment of a non-human primate model for tuberculosis allows one to obtain tissue not generally available from human studies, including lung tissue and granulomas at various times post-infection. Monkeys in these studies will be used to study localized immune responses to tuberculosis, i.e. T cells, macrophages and cytokines in the pulmonary granulomas at various times post-infection, using a variety of techniques. The rhesus or cynomolgus monkey may provide an excellent model for studying tuberculosis, including protective immune responses, but also provides a model for studying the interaction of two important pathogens: M. tuberculosis and HIV. Understanding this interaction, including the immunologic responses that occur during co-infection is key to the eventual control of these diseases. We will co-infect monkeys with SIV and M. tuberculosis to determine the effect that tuberculosis, and the chronic immune stimulation that accompanies it, has on SIV infection and progression to AIDS. Various immune parameters of localized responses in the co-infected monkeys will also be tested. Once a non-human primate model for tuberculosis is established, it could be used to test vaccines, immunotherapy, and new treatment options for tuberculosis.